I build software where the stakes are real β systems used by first responders and public safety agencies, where downtime isn't just an incident and a broken auth flow isn't just a UX bug.
I work at the intersection of Agentic AI, Identity & Security, and Full Stack Engineering. Currently deep in the MCP and agentic AI space, designing full agentic stacks and rethinking how AI agents should be secured and trusted in enterprise environments.
- π€ Building a greenfield agentic AI assistant for public safety - full stack from MCP server + orchestrator client to conversational UI
- π Leading technical design of FIDO-based MFA, phishing-resistant auth for enterprise
- π Mentoring grad students at UTD on MCP & Agentic AI workflows
- AI agents that are secure, scoped, and trusted - not just capable
- Knowing why something should or shouldn't be built as much as how
- Building in the open and giving back to the community that shapes this technology


